The Sheepskin Effect and Student Achievement

De-emphasizing the Role of Masters Degrees in Teacher Compensation

By Raegen Miller and Marguerite Roza July 2012

Beginning this year with its 2012 graduating class, the University of Notre Dame ended its practice of o ering diplomas made of sheeps skin, a tradition that has all but disappeared except in some stubborn corners of academia.1 But the tendency of employers to pay premiums to workers holding certain diplomas is thriving. is tendency, dubbed the sheepskin e ect,2 makes a labor market more e cient if those workers holding the sheepskin are indeed more productive than those without them. Most certainly, the U.S. teacher labor market could be more e cient. Although teachers with masters degrees generally earn additional salary or stipends the so-called masters bump they are no more e ective, on average, than their counterparts without masters degrees.3 e more nuanced evidence suggests that masters degrees in math and science do confer an instructional advantage on teachers of those subjects, yet approximately 90 percent of the masters degrees held by teachers come from education programs that tend to be unrelated to or unconcerned with instructional e cacy.4 A few years ago we conducted a state-by-state analysis of the nations investments in masters bumps. In that initial study we found that during the 2003-04 school year, nearly $8.6 billion was tied up nationwide in this form of teacher compensation.5 Now weve updated our analysis with data from the 2007-08 school year, the most recent data available to support such work. (see Appendix) In just four years the nations annual outlay for masters bumps surged by 72 percent to $14.8 billion. is increase, which outstripped in ation many times over during the same time period,6 is music to the ears of those institutions of higher education that cater to teachers and their academic pursuits. But for the nations primary and secondary schools, this increase strikes a discordant note and underscores the need to uncouple teacher compensation from the earning of advanced degrees.

Center for American Progress | The Sheepskin Effect and Student Achievement

e Center for American Progress and the Center on Reinventing Public Education have previously pointed out the potential advantages of more complex teacher compensation systems, in which higher pay goes to teachers in shortage subject areas, to e ective teachers who support novices or tackle the most challenging assignments, and to teachers with extraordinary instructional impact.7 is brief dissects the nations sizeable investment in masters bumps as a means of highlighting policy obstacles to a more smartly di erentiated compensation approach. We follow our recommendations with a look at some encouraging developments while at the same time de ating a canard currently misinforming reform debates. First, however, we re-examine conventional wisdom underlying the status quo in teacher compensation.

Outdated traditional teacher compensation

Nearly everything about schools in the United States from the way theyre funded to the way student achievement is measured is complicated. e big exception is teacher compensation, which is uncomplicated, but not in a good way. Scores of school districts have taken strides toward sensibly di erentiating teachers pay, o en with the catalytic support of philanthropies or the Teacher Incentive Fund,8 a competitive federal program rst funded by a 2006 appropriations bill. Yet most of the nations school districts remain shackled to the traditional, simplistic salary schedule in which just two measures ma er: years on the job and advanced degree a ainment. e rst of the traditional drivers of teacher pay is longevity or time on the job. is tradition makes sense insofar as novice teachers face a steep learning curve and early-career teachers rising to the challenge should enjoy pay increases recognizing their success just as do their counterparts in other professions.9 But teachers performance, as measured by value-added estimates of their impact on student achievement, tends to a en out a er 6 to 10 years.10 e tradition of experience-based salary increases for veteran teachers is, therefore, indi erent to student achievement. Post-baccalaureate a ainment is the other traditional driver of teacher pay, and stipends or salary di erentials tied to masters and other advanced degrees play a prominent role in this context. Masters bumps absorb a much smaller share of compensation than experience-based di erentials. is fact alone makes masters bumps the prong of traditional teacher compensation that is most amenable to reform reform that is more than justi ed by the extensive literature documenting that masters degrees do not necessarily identify e ective teachers.11 Yet at rst blush its still more than a bit counterintuitive that teachers holding a masters degree in education tend not to be especially good, relative to teachers without such credentials, at boosting student achievement. But theres no mystery to this consistent

Center for American Progress | The Sheepskin Effect and Student Achievement

research nding when one considers the substance and standards of many teacher education programs and the general academic skills of the teachers enrolled in those programs.12 Instructional e cacy is not the focus of many masters degree programs in education. Approximately 10 percent of the masters degrees held by teachers are geared toward educational administration.13 Further, some masters programs double as teacher education programs with curricula that are a confusing patchwork lacking in rigor and o en absent coursework that a reasonable person might imagine fundamental.14 e National Council on Teacher Quality found, for example, that only 15 percent of the education schools in a representative sample provided prospective teachers with even minimal exposure to the science of reading.15 In fairness, this nding pertains mainly to the undergraduate education programs that prepare approximately 70 percent of the countrys elementary school teachers.16 Yet this fact highlights another explanation for the impotency of masters degrees in education. e undergraduate education major is a consciously unselective course of study,17 a reality re ected in the low bars set by state licensure examinations.18 And while aspiring secondary teachers have SAT scores that are around average for college seniors and recent graduates, the test scores of newly minted elementary teachers lag by nearly 100 points.19 Perhaps most importantly, given teachings di culty and gravity, only 23 percent of teachers come from the top third of their graduating class.20

Poor rate of return on masters bumps

Not only does the annual outlay for masters bumps in ate demand for masters degrees, it understates the full nancial and social cost of this traditional facet of teacher compensation in the following three ways: First, the extra cost is a lost opportunity. e billions of dollars tied up in masters bumps are not available for compensation vehicles be er aligned with a school districts strategic goals such as improving student achievement. Second, some school districts o er tuition reimbursement to teachers pursuing a masters degree.21 ird, many teachers leave the classroom years before earning enough additional compensation by way of masters bumps to pay down loans or defray other expenses associated with their e orts to earn a masters degree. At this point it seems both fair and important to investigate ways to mitigate the costs of masters bumps, falling as they do on students, school districts, and teachers. A state-bystate breakdown of the main cost the masters bumps themselves helps point the way.

Center for American Progress | The Sheepskin Effect and Student Achievement

Table 1 shows by individual state and the District of Columbia the average masters bump during the 2007-08 school year, total of current expenditures devoted to this form of compensation, and what these expenditures mean on a per-pupil basis. ese dollar amounts allow o cials and policymakers from a given state to gauge their investment in masters bumps relative to the states other spending gures in education or for other public services for the 2007-08 school year. States and the District of Columbias average masters bump teacher pay increase, total expenditures on such compensation, and expenditure per pupil linked to the masters bump for the 2007-08 school year in dollars
State Alabama Alaska Arizona Arkansas California Colorado Connecticut Delaware District of Columbia Florida Georgia Hawaii Idaho Illinois Indiana Iowa Kansas Kentucky Louisiana Maine Maryland Massachusetts Michigan Minnesota Mississippi Missouri Montana Nebraska Nevada New Hampshire Average masters bump in dollars 6,030 4,840 3,040 3,970 5,890 8,010 5,906 6,230 11,280 2,850 6,880 4,524 3,730 11,910 3,830 4,160 5,520 4,570 4,810 2,940 2,080 4,890 7,600 10,090 4,800 6,180 7,340 3,290 5,810 4,890 Total funds tied up in masters bump in dollars 178,895,561 17,152,272 102,929,789 58,803,479 863,154,237 229,226,490 239,265,948 31,866,301 29,101,443 197,352,532 513,017,279 30,702,812 20,530,723 941,356,284 164,031,621 66,297,572 97,691,014 160,628,861 64,975,475 23,865,079 71,460,647 272,796,897 468,845,456 377,087,017 73,938,605 239,221,776 34,688,217 35,750,582 80,444,533 43,110,192 Expenditures per pupil tied up in bump in dollars 240 131 95 123 136 286 419 260 371 74 311 171 75 446 157 137 209 241 95 122 84 283 277 450 150 261 243 123 187 215

Center for American Progress | The Sheepskin Effect and Student Achievement

New Jersey New Mexico New York North Carolina North Dakota Ohio Oklahoma Oregon Pennsylvania Rhode Island South Carolina South Dakota Tennessee Texas Utah Vermont Virginia Washington West Virginia Wisconsin Wyoming Total

5,090 4,590 7,426 5,020 8,550 8,760 2,460 2,450 7,220 8,500 5,320 5,250 2,720 3,390 2,010 6,440 3,290 5,000 3,050 5,990 5,050

280,318,122 48,960,564 1,493,627,786 170,569,896 24,270,562 801,281,161 38,277,952 48,922,436 540,618,348 62,244,776 154,187,168 18,483,967 100,583,796 345,557,328 21,295,794 37,813,798 131,950,610 199,381,622 42,269,732 231,837,898 17,645,951

203 149 540 115 255 439 60 86 300 422 216 152 104 74 37 402 107 194 150 265 204 $14,820,002,451

Source: Authors computations using data from: National Center for Education Statistics, 2007-08 Schools and Sta ng Survey (Department of Education, 2009).

e gures in Table 1 are not suitable for comparison between states or for assessing statistical relationships among them because the costs of providing public education, especially prevailing wages, vary substantially. Table 2 shows the average masters bump and the statewide per-pupil expenditure in terms of cost-adjusted dollars. e remaining measures, inherently comparable, include the percentage of total education expenditures devoted to additional compensation for masters degrees, and the percentage of teachers holding a masters degree or higher.

Center for American Progress | The Sheepskin Effect and Student Achievement

Cost-adjusted dollar amounts and percentages characterizing states devotion to masters degrees for teachers
Average masters bump in cost-adjusted dollars* 6,591 4,959 3,150 4,543 5,130 8,112 5,285 5,853 8,793 2,945 6,729 4,579 4,321 10,859 4,180 4,757 6,320 5,019 5,199 3,368 1,816 4,359 7,691 10,144 5,399 6,610 9,161 3,820 5,555 5,054 4,348 4,950 6,352 5,148 10,077 8,855 2,835 2,585 Per-pupil expenditure for masters bumps in cost-adjusted dollars* 263 134 98 140 119 290 375 244 289 76 304 173 87 406 171 156 239 265 103 139 74 253 280 453 168 279 303 143 179 222 173 160 462 117 301 443 69 91 Percentage of total education expenditures devoted to masters bumps 2.6 0.9 1.2 1.4 1.4 3.1 2.9 2.1 2.3 0.8 3.2 1.4 1.1 4.3 1.8 1.5 2.1 2.8 1.0 1.0 0.6 2.1 2.7 4.5 1.9 2.8 2.5 1.2 2.3 1.8 1.2 1.6 3.2 1.5 2.7 4.2 0.8 0.9 Percentage of teachers with masters or above 56 44 51 41 47 57 81 62 59 39 61 53 34 55 63 40 47 79 28 46 57 69 63 58 43 53 37 47 59 51 44 47 88 35 32 68 33 63

State Alabama Alaska Arizona Arkansas California Colorado Connecticut Delaware District of Columbia Florida Georgia Hawaii Idaho Illinois Indiana Iowa Kansas Kentucky Louisiana Maine Maryland Massachusetts Michigan Minnesota Mississippi Missouri Montana Nebraska Nevada New Hampshire New Jersey New Mexico New York North Carolina North Dakota Ohio Oklahoma Oregon

Center for American Progress | The Sheepskin Effect and Student Achievement

Pennsylvania Rhode Island South Carolina South Dakota Tennessee Texas Utah Vermont Virginia Washington West Virginia Wisconsin Wyoming

7,151 8,133 5,695 6,537 2,864 3,300 2,110 7,223 2,913 4,649 3,435 6,146 5,730

297 403 232 189 110 72 39 451 95 180 168 272 232

2.6 2.9 2.4 1.8 1.3 0.9 0.6 2.8 1.0 2.1 1.5 2.5 1.5

55 55 59 33 55 30 39 57 43 69 61 55 44

* Cost adjustment made using 2008 state version of the Comparable Wage Index, or CWI, created by Lori Taylor, Bush School of Government and Public Service, Texas A&M University. The 2008 Comparable Wage Index is neither published nor endorsed by the National Center for Education Statistics, but the methodology underlying it is that of the 19972004 versions of the Comparable Wage Index that were published by the National Center for Education Statistics.21 Source: Authors computations using data from: National Center for Education Statistics, 2007-08 Schools and Sta ng Survey (Department of Education, 2009).

So what do we make of all this? First, there is a strong relationship between the average masters bump and the share of expenditures directed toward them.23 is relationship is no surprise because masters stipends and salary di erentials are fundamental cost drivers at the district level one would need masters bump information to predict expenditures. Yet the share of expenditures tied to masters degrees is almost never the starting point or even on the table during conversations about altering compensation systems. In an era of especially scarce resources, it is certainly a discussion worth having. e per-pupil expenditure gures represent another potent, though still exotic, way of portraying states compensation priorities. Moreover, the tremendous range in values $39 per pupil tied to the masters bump in Utah to $462 in New York o ers a gauge of the di erences between states teacher labor markets. In particular, the expenditure-per-pupil range re ects the drastically di erent probabilities of Louisiana and New York students being taught by a teacher with a masters degree. ese probabilities, represented by the percentage of teachers with a masters or above, cut close to policies shaping traditional teacher compensation. Casual economic thinking might lead one to imagine a strong statistical relationship between the size of the typical masters bump in a state and the percentage of teachers being compensated in this way. is is not the case as the following examples illustrate.24 e top ve states in terms of the percentage of teachers holding masters degrees are New York (88 percent), Connecticut (81 percent), Kentucky (79 percent), Massachuse s (69 percent), and Washington (69 percent), yet their average masters bumps fall between $4,649 and $6,352, in the middle of the pack. Meanwhile, Montana and North Dakota have two of the lowest percentages of teachers holding advanced

Center for American Progress | The Sheepskin Effect and Student Achievement

degrees (37 percent and 32 percent, respectively) while boasting masters bumps among the highest ve in the country $9,161 for Montana and $10,077 for North Dakota.

State policies that matter

Two state-level policies have more to do with the percentage of teachers holding a masters degree than does the average pay di erential tied to the degree. Eight states make holding an advanced degree a condition for receiving a professional license, as opposed to a probationary or provisional one, and 16 states require employers to pay teachers more if they hold advanced degrees. Table 3 showcases the states espousing either of these policies, in descending order by the percentage of teachers with advanced degrees. Among these states, the average percentage of teachers with an advanced degree is 57 percent; among those with neither policy, its 48 percent. States with policies that promote teachers acquisition of masters degrees
State New York Connecticut Kentucky Washington Ohio Oregon Michigan Delaware Georgia West Virginia South Carolina Maryland Alabama Tennessee Illinois Hawaii Mississippi Arkansas Montana North Carolina Oklahoma Louisiana Total 16 8
Source: National Council on Teacher Quality, 2011 State Teacher Policy Yearbook (2012), available at http://www.nctq.org/stpy11Home.do.

Require extra pay for advanced degree

Require advanced degree for full professional license

Percentage of teachers with advanced degree 88 81 79 69 68 63 63 62 61 61 59 57 56 55 55 53 43 41 37 35 33 28

Center for American Progress | The Sheepskin Effect and Student Achievement

e policy perspective certainly makes it easier to understand why teachers in states such as Maryland and Tennessee have advanced degrees more o en than not despite their states relatively low spending for masters bumps. At the microeconomic level, where individual teachers weigh their options, the magnitude of nancial incentives tied to a masters degree clearly ma ers, but these state policies ma er too. is brings us to our policy recommendations.

Make masters programs compete on merit

Certainly it is true that prospective teachers with a bachelors degree need some additional speci c preparation for teaching. Likewise, those experienced teachers remaining in the profession need feedback on their performance and opportunities to improve their practice. And arguably, masters degrees may have a role to play in those instances, but state policymakers should dispense with policies that mandate di erential pay for teachers with advanced degrees or that make advanced degrees a requirement for remaining in the profession. ese policies heed a conventional wisdom thats oblivious to strategic concerns around bolstering the quality of the teacher workforce, improving student outcomes overall, and closing achievement gaps between groups of students de ned by ethnicity or economic status. ese changes alone may not have a huge or immediate impact on teacher compensation systems, but they will enable local policymakers to begin de-emphasizing a traditional driver of teacher compensation the advanced degree. e masters bump in many school districts takes the form of an annual stipend si ing on top of a teachers salary. Rather than increasing such stipends in conjunction with cost of living increases to salary, which is a common practice, districts could and should avoid directing new resources toward them. In other districts the masters bump has penetrated the salary schedule. Merging the salary columns for teachers with and without masters degrees in some type of buyout approach would likely be cost prohibitive or simply imprudent. It may be possible, however, for districts to create di erent salary schedules for new teaching hires that are neutral with respect to masters degrees while grandfathering the masters bumps of existing teachers. Undoubtedly, moves to de-emphasize the role of masters degrees in teacher compensation will run into opposition. A hollow but fashionable argument in support of masters degrees arises from international comparisons used to inform current debates about education reform. Lets turn to this argument now.

Center for American Progress | The Sheepskin Effect and Student Achievement

Finland-topia
All teachers in Finland have a masters degree, and they get extraordinary results from their students. If we want be er results in U.S. schools, then, we should require teachers to have a masters degree, so the argument goes. But this argument has two fatal aws. First, teachers in Finland hail from the top 10 percent of their graduating class.25 is selectivity is woven into a set of policies that Linda Darling-Hammond, a professor at Stanford University, has astutely described as a teaching and learning system.26 e Finnish system could scarcely be more di erent than our domestic grab bag of policies arising from approximately 15,000 separate school districts carrying out their responsibility to provide public education, variously conceived by the diverse states of a country with an unmatched tolerance, at least among wealthy industrialized nations, for inequity in school funding and facilities. Secondly, Finnish teachers hold masters degrees that augment their knowledge and skills in a way thats deliberately connected to their instructional challenges. Secondary teachers earn a masters in the subject of instruction, and the masters degree required of elementary teachers equips them with specialized knowledge and skills o en found only among special education teachers and school psychologists in U.S. schools. us, holding masters degrees means Finnish teachers either have a serious grasp on academic content or are well equipped to problem solve around the individual learning needs of their students. e typical masters degree held by a U.S. teacher and the associated skills a ached pale in comparison. Moreover, its unlikely to move in this direction barring a tectonic shi in the higher-education landscape. Institutions of higher education, of course, wont be at the vanguard of e orts to repeal legislation that in ates demand for one of their most lucrative products masters degrees in education. In addition, it bears mentioning, for example, that Connecticuts requirement that teachers seeking a professional license hold a masters degree was unscathed by recent reform-conscious legislation in that state.

Hopeful signs
Schools of education, many reeling from the e ects of the economic recession, can proactively begin tuning themselves up to compete for the tuition dollars of teachers and would-be teachers. is would certainly be smart business, particularly if the writing is indeed on the wall for licensure rules and traditional compensation that favor masters degrees irrespective of their strategic merit.

10

Center for American Progress | The Sheepskin Effect and Student Achievement

e con uence of two movements shows that many schools of education are at least taking note. e rst movement revolves around the idea of performance assessment.27 Consortia of education schools have begun taking stock of their masters candidates actual performance with respect to standards describing e ective teaching practice. Performance assessment schemes at some schools may not pay quite enough a ention to student achievement for some tastes, but the Relay Graduate School of Education in New York City is pushing the performance assessment envelope.28 ose teachers earning a masters degree from Relay must demonstrate at least a years worth of academic growth in their students where this can be measured using test scores or in 70 percent mastery of academic standards, in other cases.29 e second movement is the overhaul of school districts practices around performance evaluation of their largest employee group teachers. is movement, propelled by competitive federal grant programs such as Race to the Top, faces all manner of implementation challenges, but states and districts should wind up with useful data systems and much improved practices around teacher evaluation. Of course, they couldnt do worse than the infamous status quo in which 99 percent of teachers receive satisfactory marks.30 ose teachers seeking masters degrees should have access to appropriately aggregated information about performance assessment results of past candidates and performance evaluation results of graduates. Such information would enable teachers and prospective teachers to select programs to optimize their chances of securing pay di erentials tied to performance or the di culty of their assignments. Louisiana remains in the lead among states in developing a system of accountability for teacher preparation programs31 a promising policy vehicle for creating a marketplace for information about program e cacy.

A broad imperative for masters bump divestment

An independent taskforce report published recently by the Council on Foreign Relations warns, Educational failure puts the United States future economic prosperity, global position, and physical safety at risk.32 Failure in the form of inadequate student achievement and achievement gaps, of course, has many causes, but traditional teacher compensation systems are undeniably part of the problem. Teachers are the most important school-based resource a ecting student achievement,33 and the lions share of school spending goes toward the nancial compensation for teachers.34 Policymakers wishing to take steps toward smartly di erentiated compensation for teachers have to start somewhere. Divesting in masters bumps by following the discrete recommendations weve o ered may be one of the easier places to start. e disconnect between the goal of improving student achievement and the tradition of paying teachers extra simply for holding post-baccalaureate sheepskin certainly makes doing so strategically defensible.

11

Center for American Progress | The Sheepskin Effect and Student Achievement

Appendix
is analysis used data from the 2007-08 Schools and Sta ng Survey from the National Center for Education Statistics that provided state-by-state gures for both the percentage of teachers with masters degrees and the average salary of teachers at each degree level (bachelors or below, masters, etc.) for given years of longevity. is analysis used these data to compute the average percentage salary increase awarded for education credits earned beyond a bachelors degree. e analysis then applied the percentage increases to the more recent state-by-state average salary gures and total number of teachers from the National Educators Associations 2008-09 Salary Survey in order to compute the dollar value of the masters bump in each state. As reported here, the dollar increase on the salary for a masters degree is the average di erence between the salary for a teacher with a bachelors degree (with no extra credits) and the salary for a teacher with a masters degree for a given experience level. In other words, this bump includes all salary increments for credits earned for any level of education beyond the bachelors degree. Finally, these salary bumps do not include any amounts districts spent on subsidizing teachers costs for earning higher degrees. Raegen Miller is the Associate Director for Education Research at the Center for American Progress. Marguerite Roza is a research associate professor at the University of Washingtons College of Education and senior scholar at the Center on Reinventing Public Education.

Endnotes
1 Ronnie Reese, Notre Dame ditching sheepskin diplomas, Chicago Tribune, October 12, 2011, available at http://articles.chicagotribune.com/2011-10-12/news/ct-talk-sheepskin-diplomas-1012-20111012_1_sheepskins-graduatesparchment. 2 The sheepskin e ect emanates from Michael Spences seminal work on labor-market signaling. See: Michael Spence, Job Market Signaling, Quarterly Journal of Economics 87 (3) (1973): 355374. 3 Dan D. Goldhaber, Dominic J. Brewer, and Deborah J. Anderson, A Three-Way Error Components Analysis of Educational Productivity, Education Economics 7 (3) (1999); Thomas J. Kane, Jonah E. Rocko , and Douglas O. Staiger, What Does Certi cation Tell Us About Teacher E ectiveness? Evidence from New York City (Cambridge: National Bureau of Economic Research, 2006); Daniel Aaronson, Lisa Barrow, and William Sander, Teachers and Student Achievement in the Chicago Public High Schools. Working Paper WP-02-28 (Federal Reserve Bank of Chicago, 2002); Eric A. Hanushek, John F. Kain, and Steven G. Rivkin, Teachers, Schools, and Academic Achievement. Working Paper 6691 (National Bureau of Economic Research, 1998); National Council on Teacher Quality, Increasing the Odds: How Good Policies Can Yield Better Teachers (2005). 4 National Center for Education Statistics, 2003-04 Schools and Sta ng Survey. 5 Marguerite Roza and Raegen Miller, Separation of Degrees (Washington: Center for American Progress, 2009), available at http://www.americanprogress.org/issues/2009/07/separation_of_degrees.html. 6 The Consumer Price Index, or CPI-U, increased by just 13.1 percent from 2004 to 2008. See: Consumer Price Index, available at ftp://ftp.bls.gov/pub/special.requests/cpi/cpiai.txt. 7 Daniel Goldhaber, Teacher Pay Reforms (Washington: Center for American Progress, 2006), available at http:// www.americanprogress.org/issues/2006/12/teacher_pay. html; Michael DeArmond and Daniel Goldhaber, A Leap of Faith: Redesigning Teacher Compensation (Seattle: Center on Reinventing Public Education, 2008), available at http:// www.crpe.org/cs/crpe/view/csr_pubs/162. 8 The Rose Community Foundations $4 million investment in Denvers ProComp is an example of an early and large philanthropic investment in di erentiated compensation. See: Education, available at http://www.rcfdenver.org/ programs_education.htm. See also: Robin Chait and Raegen Miller, Teacher Incentive Fund Addresses Three Key Issues (Washington: Center for American Progress, 2009), available at http://www.americanprogress.org/issues/2009/06/ teacher_incentive_fund.html. 9 Jacob Vigdor, Scrap the Sacrosanct Salary Schedule, Education Next 8 (4) (2008), available at http://educationnext.org/ scrap-the-sacrosanct-salary-schedule/.

12

Center for American Progress | The Sheepskin Effect and Student Achievement

10 Hanushek, Kain, and Rivkin, Teachers, Schools, and Academic Achievement; Rocko , The Impact of Individual Teachers on Student Achievement: Evidence from Panel Data; Douglas O. Staiger, Robert Gordon, and Thomas J. Kane, Identifying E ective Teachers Using Performance on the Job (Washington: Brookings Institution, 2006); Raegen T. Miller, Richard J. Murnane, and John B. Willett, Do Teacher Absences Impact Student Achievement? Longitudinal Evidence from One Urban School District, Educational Evaluation and Policy Analysis 30 (2) (2008). 11 Goldhaber, Brewer, and Anderson, A Three-Way Error Components Analysis of Educational Productivity; Kane, Rocko , and Staiger, What Does Certi cation Tell Us About Teacher E ectiveness? Evidence from New York City; Aaronson, Barrow, and Sander, Teachers and Student Achievement in the Chicago Public High Schools; Hanushek, Kain, and Rivkin, Teachers, Schools, and Academic Achievement; National Council on Teacher Quality, Increasing the Odds: How Good Policies Can Yield Better Teachers. 12 Representatives from the masters degree industry often criticize this research base, especially its focus on measures of student achievement. For a classic example, see: Allie Graspgreen, Questioning a Degrees Value, Inside Higher Ed, December 9, 2010, available at http://www.insidehighered.com/news/2010/12/09/masters. 13 Our analyses reveal that 53 percent of teachers hold a masters degree, and 5.6 percent of all teachers hold such degrees in administration. See: Number of public school teachers and number and percentage of public school teachers who held a masters degree or higher in Administration, by selected school characteristics: 200708, available at http://nces.ed.gov/surveys/sass/tables/ sass0708_031_t1n.asp. 14 Arthur Levine, Educating School Teachers (Washington: The Education Schools Project, 2006), available at http:// www.edschools.org/pdf/Educating_Teachers_Report.pdf. 15 National Council on Teacher Quality, What Education Schools Arent Teaching about Reading and What Elementary Teachers Arent Learning (2007), available at, http:// www.nctq.org/p/publications/docs/nctq_reading_study_ app_20071202065019.pdf. 16 Julie Greenberg and Kate Walsh, No Common Denominator: The Preparation of Elementary Teachers in Mathematics by Americas Education Schools (Washington: National Council on Teacher Quality, 2008), available at http://www. nctq.org/p/publications/docs/nctq_ttmath_fullreport.pdf. 17 The most selective teacher education programs in the country are less likely to be accredited by the National Council for Accreditation of Teacher Education. See: Levine, Educating School Teachers. 18 National Council on Teacher Quality, 2011 State Teacher Policy Yearbook (2011), available at http://www.nctq.org/ stpy11AreaScores.do?id=6. 19 Levine, Educating School Teachers. 20 Byron Auguste, Paul Kihn, Matt Miller, Closing the talent gap: Attracting and retaining top third graduates to a career in teaching (McKinsey & Company, 2010), available at http://mckinseyonsociety.com/closing-the-talent-gap/.

21 Chad Aldeman, The Condition of Education: Masters Degrees in Education, The Quick and the Ed Blog, June 1, 2009, available at http://www.quickanded.com/2009/06/ condition-of-education-masters-degrees.html. 22 Lori L. Taylor and Mark C. Glander, Documentation for the NCES Comparable Wage Index Data File (Department of Education, 2006). We obtained the 2008 CWI index from Lori Taylor. The index is constructed in the same manner as the NCES version, which is not available for years after 2005. 23 The correlation between average masters bump and percentage of expenditures devoted to masters bumps is .85. 24 The correlation between the percentage of teachers with an advanced degree and the average masters bump is approximately .16. 25 Strong Performers and Successful Reformers in Education: Finland, available at http://www.pearsonfoundation.org/ oecd/ nland.html. 26 For a thorough examination of the contrast between the education systems, especially in teacher policy, of Finland and the United States, see: Linda Darling-Hammond, What we can learn from Finlands successful school reform, NEA Today, October 2010, available at http://www.nea.org/ home/40991.htm. 27 Linda Darling-Hammond, Evaluating Teacher E ectiveness: How Teacher Performance Assessments Can Measure and Improve Teaching (Washington: Center for American Progress, 2010), available at http://www.americanprogress. org/issues/2010/10/teacher_e ectiveness.html. 28 ConnCAN, 50CAN, and Public Impact, Measuring Teacher E ectiveness: A Look Under the Hood of Teacher Evaluation in 10 Sites (2012), available at http://conncan.org/sites/ conncan.org/ les/research/measuring_teacher_e ectiveness.pdf. 29 June Kronholz, A New Type of Ed School: Linking candidate success to student success, Education Next 12 (2) (2012). 30 Daniel Weisberg and others, The Widget E ect (New York: The New Teacher Project, 2009), available at http://widgeteffect.org/. 31 What Louisiana Can Teach, The New York Times, December 12, 2008, available at http://www.nytimes.com/2008/12/12/ opinion/12fri2.html?_r=1. 32 Council on Foreign Relations, U.S. Educational Reform and National Security (2012), available at http://www.cfr. org/united-states/us-education-reform-national-security/ p27618. 33 Staiger, Gordon, and Kane, Identifying E ective Teachers Using Performance on the Job. 34 Karen Hawley Miles and Stephen Frank, The Strategic School: Making the Most of People, Time, and Money (Thousand Oaks, CA: Corwin Press, 2008).

13

Center for American Progress | The Sheepskin Effect and Student Achievement